On Fri, 2003-02-21 at 19:48, Adam Williamson wrote: > On Fri, 2003-02-21 at 22:45, Buchan Milne wrote: > > > Do not know. I am using acpi on my IBM 600X, which seems to work ok, > > except that I do not see any benefit over APM (bseides the fact that I can > > view temperatures and charge in mWh). Am I supposed to see my machine > > suspend automatically (it does not)? All my normal devices (PCMCIA NIC, > > winmodem, USB port etc) work, but I was hoping for a bit more ... >
If you had a modern laptop, you'd be happy to get that because it wouldn't happen without ACPI :-) > No. That's what acpid is for. It can trigger ACPI events when certain > things happen, as apmd (allegedly...) can for APM. This presupposes you > can get suspending to work at all, of course. :) > > > Maybe I should go back to apm, then at least I can use apmiser to throttle > > the CPU (which I am not sure acpi does at present). > If your hardware supports APM, I'd use APM until the 2.6 kernel is available. That's where the real ACPI development will take place, AFAICT everything in 2.4 is backports that seem to be done as afterthoughts. > It ought to be possible. Investigate > /proc/acpi/processor/CPU0/throttling . It is possible, but you'll want a daemon to maximize usage rather than echo-ing values into /proc. See autospeedstep or cpufreq. > > I get the feeling if someone would just write some damn *documentation* > for the current ACPI implementation we'd find out all sorts of cool > things it does. As it stands, anyone who isn't a developer is fumbling > around in the dark... seriously. The ACPI support mailing list is occasionally illuminating, the swsusp mailing list more so. -- Jack Coates Monkeynoodle: A Scientific Venture...
